GEOGRAPHICAL RESEARCH ›› 2016, Vol. 35 ›› Issue (12): 2298-2308.doi: 10.11821/dlyj201612009

• Articles • Previous Articles     Next Articles

Measurement and spatial analysis of poverty-strickenvillages in China

Yefeng CHEN1,2,3(), Yanhui WANG1,2,3(), Xiaolin WANG4   

  1. 1. Beijing Key Laboratory of Resource Environment and Geographic Information System, Capital Normal University, Beijing 100048, China
    2. Key Laboratory of 3-Dimensional Information Acquisition and Application, Ministry of Education, Capital Normal University, Beijing 100048, China
    3. State Key Laboratory Incubation Base of Urban Environmental Processes and Digital Simulation, CapitalNormal University, Beijing 100048, China
    4. Information Center of the State Council LeadingGroup Office of Poverty Alleviation and Development, Beijing 100028, China
  • Received:2016-06-05 Revised:2016-09-22 Online:2016-12-23 Published:2016-12-23


Rural poverty is one of the most important obstacles to achieve the goal that builds a moderately prosperous society in an all-around way in China. To address the strategic requirement of targeted poverty reduction for the implementation of "Entire-village Advancement" to rid all Chinese of poverty by 2020, we integrate geographical contributing factors and socioeconomic ones to design a set of village-level evaluation indicator system, consisting of 6 dimensions and 20 indicators. And then based on the above indicator system, we build a comprehensive village-level poverty measurement model, using village-level archived "Entire-village Advancement" dataset that was collected during the "12th Five-year Plan" period (from 2011 to 2015) to assess multidimensional poverty index and relative poverty characteristics of each village. At last, we adopt spatial autocorrelation and weighted kernel density estimation to examine spatial distribution of the poverty-stricken villages at a multiscale of area-province-county. The results show that: (1) During the "12th Five-year Plan" period, the distribution of poverty levels for different villages statistically represents a geometrical olive-shape pattern that is larger in the middle and smaller at both ends; Meanwhile, the poverty depths of poverty-stricken villages at both provincial and county scales are closely related to their local economies, regional locations, local policies and natural environment, etc. (2) As far as the distribution of poverty-stricken villages is concerned, there exists a spatially heterogeneous distribution, presenting a typical sandwich-shaped structure that is sparse in both eastern and northwestern China while dense in both central and southwestern China, where different levels of dotted poverty kernels are scattered, and 3 first-level kernels, 6 secondary kernels and many tertiary kernels of poverty hotspots exist obviously. However, a grey poverty area called "transition zone", located between High-High and Low-Low poverty-stricken areas, is found in central China where the distribution of poverty-stricken villages does not show a significant spatial autocorrelation. (3) The multidimensional characteristics of poverty-stricken villages represent a globally strong spatial dependence with a Moran's I coefficient of 0. 55; however, both High-High areas and Low-Low areas are distributed intensively while both High-Low areas and Low-High areas are distributed discretely, overall showing a stepped distribution pattern that is "west-high vs. east-low". This study could help precisely target the poverty situation of the poverty-stricken villages in rural China, and also may provide a good understanding of the status and regional differences among villages at a village scale. On the other hand, it could serve as a scientific reference regarding decisions-making in both promoting intra-rural anti-poverty harmonious development and in constructing the new countryside of China.

Key words: poverty-stricken villages, GIS, multidimensional poverty index, spatial autocorrelation, Weighted Kernel Density Estimation